Parsimonious Morpheme Segmentation with an Application to Enriching Word Embeddings

TL;DR

This paper introduces MorphMine, an unsupervised method for morpheme segmentation that improves word embeddings by leveraging subword structures, leading to better semantic representations especially for rare and out-of-vocabulary words.

Contribution

MorphMine is a novel, parsimonious hierarchical morpheme segmentation method that enhances word embedding quality across multiple languages and tasks.

Findings

MorphMine segments words into human-verified morphemes across languages.

Enriching embeddings with MorphMine morphemes improves performance on evaluation tasks.

The method effectively handles infrequent and out-of-vocabulary words.

Abstract

Traditionally, many text-mining tasks treat individual word-tokens as the finest meaningful semantic granularity. However, in many languages and specialized corpora, words are composed by concatenating semantically meaningful subword structures. Word-level analysis cannot leverage the semantic information present in such subword structures. With regard to word embedding techniques, this leads to not only poor embeddings for infrequent words in long-tailed text corpora but also weak capabilities for handling out-of-vocabulary words. In this paper we propose MorphMine for unsupervised morpheme segmentation. MorphMine applies a parsimony criterion to hierarchically segment words into the fewest number of morphemes at each level of the hierarchy. This leads to longer shared morphemes at each level of segmentation. Experiments show that MorphMine segments words in a variety of languages into human-verified morphemes. Additionally, we experimentally demonstrate that utilizing MorphMine morphemes to enrich word embeddings consistently improves embedding quality on a variety of of embedding evaluations and a downstream language modeling task.